<div dir="ltr"><div><div><div><div>Hi,<br><br></div>We ran a test on GlusterFS 3.12.1
with erasurecoded volumes 8+2 with 10 bricks (default config,tested with
100gb, 200gb, 400gb bricksizes,10gbit nics)<br><br>1.<br></div>Tests
show that healing takes about double the time on healing 200gb vs 100,
and abit under the double on 400gb vs 200gb bricksizes. Is this expected
behaviour? In light of this would make 6,4 tb bricksizes use ~ 377
hours to heal.<br></div><div><br></div><div>100gb brick heal: 18 hours (8+2)</div><div><div>200gb brick heal: 37 hours (8+2) +205%<br></div><div><div>400gb brick heal: 59 hours (8+2) +159%<br></div></div></div><div><br></div><div>Each 100gb is filled with 80000 x 10mb files (200gb is 2x and 400gb is 4x)<br></div><div><br>2.<br></div>Are
there any possibility to show the progress of a heal? As per now we run
gluster volume heal volume info, but this exit's when a brick is done
healing and when we run heal info again the command contiunes showing
gfid's until the brick is done again. This gives quite a bad picture of
the status of a heal. <br><br>3.<br></div>What kind of config tweaks is recommended for these kind of EC volumes?<br><br><br clear="all"><div>$ gluster volume info<br></div><div>Volume Name: test-ec-100g<br>Type: Disperse<br>Volume ID: 0254281d-2f6e-4ac4-a773-<wbr>2b8e0eb8ab27<br>Status: Started<br>Snapshot Count: 0<br>Number of Bricks: 1 x (8 + 2) = 10<br>Transport-type: tcp<br>Bricks:<br>Brick1: dn-304:/mnt/test-ec-100/brick<br>Brick2: dn-305:/mnt/test-ec-100/brick<br>Brick3: dn-306:/mnt/test-ec-100/brick<br>Brick4: dn-307:/mnt/test-ec-100/brick<br>Brick5: dn-308:/mnt/test-ec-100/brick<br>Brick6: dn-309:/mnt/test-ec-100/brick<br>Brick7: dn-310:/mnt/test-ec-100/brick<br>Brick8: dn-311:/mnt/test-ec-2/brick<br>Brick9: dn-312:/mnt/test-ec-100/brick<br>Brick10: dn-313:/mnt/test-ec-100/brick<br>Options Reconfigured:<br>nfs.disable: on<br>transport.address-family: inet<br> <br>Volume Name: test-ec-200<br>Type: Disperse<br>Volume ID: 2ce23e32-7086-49c5-bf0c-<wbr>7612fd7b3d5d<br>Status: Started<br>Snapshot Count: 0<br>Number of Bricks: 1 x (8 + 2) = 10<br>Transport-type: tcp<br>Bricks:<br>Brick1: dn-304:/mnt/test-ec-200/brick<br>Brick2: dn-305:/mnt/test-ec-200/brick<br>Brick3: dn-306:/mnt/test-ec-200/brick<br>Brick4: dn-307:/mnt/test-ec-200/brick<br>Brick5: dn-308:/mnt/test-ec-200/brick<br>Brick6: dn-309:/mnt/test-ec-200/brick<br>Brick7: dn-310:/mnt/test-ec-200/brick<br>Brick8: dn-311:/mnt/test-ec-200_2/<wbr>brick<br>Brick9: dn-312:/mnt/test-ec-200/brick<br>Brick10: dn-313:/mnt/test-ec-200/brick<br>Options Reconfigured:<br>nfs.disable: on<br>transport.address-family: inet</div><div><br></div>Volume Name: test-ec-400<br>Type: Disperse<br>Volume ID: fe00713a-7099-404d-ba52-<wbr>46c6b4b6ecc0<br>Status: Started<br>Snapshot Count: 0<br>Number of Bricks: 1 x (8 + 2) = 10<br>Transport-type: tcp<br>Bricks:<br>Brick1: dn-304:/mnt/test-ec-400/brick<br>Brick2: dn-305:/mnt/test-ec-400/brick<br>Brick3: dn-306:/mnt/test-ec-400/brick<br>Brick4: dn-307:/mnt/test-ec-400/brick<br>Brick5: dn-308:/mnt/test-ec-400/brick<br>Brick6: dn-309:/mnt/test-ec-400/brick<br>Brick7: dn-310:/mnt/test-ec-400/brick<br>Brick8: dn-311:/mnt/test-ec-400_2/<wbr>brick<br>Brick9: dn-312:/mnt/test-ec-400/brick<br>Brick10: dn-313:/mnt/test-ec-400/brick<br>Options Reconfigured:<br>nfs.disable: on<br>transport.address-family: inet<br clear="all"><br>-- <br><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><p></p>
Regards<br>
Rolf Arne Larsen<br>
Ops Engineer<br>
<a href="mailto:rolf@startsiden.no" target="_blank"><span>rolf@jottacloud.com</span></a><br>
<a href="http://www.jottacloud.com" target="_blank"></a></div></div></div></div>
</div>